</> Code with Papers by Syntology

Papers › Syntactically Guided Generative Embeddings for Zero-Shot Skeleton Action Recognition

Syntactically Guided Generative Embeddings for Zero-Shot Skeleton Action Recognition

27 Jan 2021arXiv:2101.11530archive 2025-07-28

Pranay Gupta, Divyanshu Sharma, Ravi Kiran Sarvadevabhatla

We introduce SynSE, a novel syntactically guided generative approach for Zero-Shot Learning (ZSL). Our end-to-end approach learns progressively refined generative embedding spaces constrained within and across the involved modalities (visual, language). The inter-modal constraints are defined between action sequence embedding and embeddings of Parts of Speech (PoS) tagged words in the corresponding action description. We deploy SynSE for the task of skeleton-based action sequence recognition. Our design choices enable SynSE to generalize compositionally, i.e., recognize sequences whose action descriptions contain words not encountered during training. We also extend our approach to the more challenging Generalized Zero-Shot Learning (GZSL) problem via a confidence-based gating mechanism. We are the first to present zero-shot skeleton action recognition results on the large-scale NTU-60 and NTU-120 skeleton action datasets with multiple splits. Our results demonstrate SynSE's state of the art performance in both ZSL and GZSL settings compared to strong baselines on the NTU-60 and NTU-120 datasets. The code and pretrained models are available at https://github.com/skelemoa/synse-zsl

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

skelemoa/synse-zsl officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Action RecognitionGeneralized Zero Shot skeletal action recognitionGeneralized Zero-Shot LearningPOSZero Shot Skeletal Action RecognitionZero-Shot Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Generalized Zero Shot skeletal action recognition NTU RGB+D SynSE Harmonic Mean (12 unseen classes) 36.33 #3 of 4 Archive leaderboard report
Generalized Zero Shot skeletal action recognition NTU RGB+D SynSE Harmonic Mean (5 unseen classes) 59.02 #3 of 4 Archive leaderboard report
Generalized Zero Shot skeletal action recognition NTU RGB+D 120 SynSE Harmonic Mean (10 unseen classes) 54.94 #3 of 4 Archive leaderboard report
Generalized Zero Shot skeletal action recognition NTU RGB+D 120 SynSE Harmonic Mean (24 unseen classes) 41.04 #3 of 4 Archive leaderboard report
Zero Shot Skeletal Action Recognition NTU RGB+D SynSE Accuracy (12 unseen classes) 33.30 #7 of 9 Archive leaderboard report
Zero Shot Skeletal Action Recognition NTU RGB+D SynSE Accuracy (5 unseen classes) 75.81 #7 of 9 Archive leaderboard report
Zero Shot Skeletal Action Recognition NTU RGB+D SynSE Random Split Accuracy 64.19 #7 of 9 Archive leaderboard report
Zero Shot Skeletal Action Recognition NTU RGB+D 120 SynSE Accuracy (10 unseen classes) 62.69 #7 of 9 Archive leaderboard report
Zero Shot Skeletal Action Recognition NTU RGB+D 120 SynSE Accuracy (24 unseen classes) 38.70 #7 of 9 Archive leaderboard report
Zero Shot Skeletal Action Recognition PKU-MMD SynSE Random Split Accuracy 53.85 #7 of 7 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections